1 $Id: CHANGES,v 1.176 2002/05/17 08:10:43 stelian Exp $
3 Changes between versions 0.4b28 and 0.4b29 (released ??????????????)
4 ====================================================================
6 1. Fixed a problem in the rmt ioctl command, where ioctl's issued from
7 non Linux clients were misinterpreted. The description of the problem
8 (incompatible numbering in Linux mtio opcodes) is documented at
9 ftp://ftp.fokus.gmd.de/pub/unix/star/README.mtio . Thanks to
10 Jörg Schilling <schilling@fokus.gmd.de> for reporting this bug and
11 providing an excellent, cross-platform replacement for rmt in his
14 2. Fixed a bug reported by Andy Mentges <amentges@jumpline.com>
15 which caused restore to fail when the inode maps were not entirely
16 contained on the first volume (when using really small volumes or
17 when dumping a huge number of inodes).
19 3. Fixed a problem in dump, where files in subdirectories of directories
20 that have the nodump flag set (or being excluded with -e/-E) were
21 sometimes incorrectly dumped (instead of being excluded). The original
22 fix is from David C Lawrence <tale@dd.org> in the FreeBSD version
23 of dump: http://www.FreeBSD.org/cgi/query-pr.cgi?pr=32414
24 Thanks to Ted Grzesik <tedgyz@roostme.com> for reporting the bug and
25 help testing the patch.
27 4. Added some example scripts from Gerd Bavendiek <bav@epost.de>
28 which makes one able to pipe the output of dump, by the net, to
29 a remote CD-burner server.
31 Changes between versions 0.4b27 and 0.4b28 (released April 12, 2002)
32 ====================================================================
34 1. Fixed a bug in the QFA feature which could have caused
35 a bad tape positionning on restore (causing extra delays in
36 restore). Credits go to Uwe Gohlke <uwe@ugsoft.de>.
38 2. Added a small note in the dump man page specifying that
39 there should be no white space between the option letter and
40 the -j or -z parameter, thanks to Kenneth Porter <shiva@well.com>
42 3. Made restore work with older versions of the readline library,
43 by checking in configure for several extended symbols. Restore
44 can now be compiled with a readline as old as the 2.0 release
45 (though it may be a good idea to upgrade it to a more recent
46 version...). Thanks to Andrew Donkin <ard@waikato.ac.nz> for
47 reporting the build failures.
49 4. Fixed a performance problem with the QFA file creation in
50 dump, which made unnecessary seeks on the tape slowing down
51 the dump. Thanks to Andrew Donkin <ard@waikato.ac.nz> for
54 5. Removed the inclusion of some kernel headers in the dump
55 source, which prevented the compile in some kernel/glibc
56 headers/architecture combination. Thanks to Bdale Garbee
57 <bdale@gag.com> for reporting the bug.
59 6. Added the appropriate error message when dump fails to
60 open the output file for writing. Thanks to Amith Varghese
61 <amithv@yahoo.com> for reporting this bug.
63 7. Made restore able to understand large Solaris ufsdump tapes
64 (containing inodes bigger than 4194304). Sun have introduced
65 an "extension" to the dump tape format when dealing with
66 those inodes, which was uncorrectly handled by Linux restore.
67 Thanks to Uwe Gohlke <uwe@ugsoft.de> for reporting the bug and
68 providing a test case.
70 8. Added the -m parameter to dump which optimises the output for
71 inodes having been changed but not modified since the last dump
72 ('changed' and 'modified' have the meaning defined in stat(2)).
73 For those inodes, dump will save only the metadata, instead of
74 saving the entire inode contents. Inodes which are either
75 directories or have been modified since the last dump are saved
76 in a regular way. Uses of this flag must be consistent, meaning
77 that either every dump in an incremental dump set have the flag,
80 Changes between versions 0.4b26 and 0.4b27 (released February 15, 2002)
81 =======================================================================
83 1. Fixed behaviour of dump when exceeding resource limits
86 2. Added the -L flag to restore to allow the user to specify a
87 maximal allowed number of miscompares when using restore
88 with the -C option to check the backup.
90 3. Detailed the manual entry for the -N option of restore.
92 4. Added the -a flag to restore to make able doing unattended
93 restores in -i or -x mode (automatically walks through the
94 multiple dump volumes).
96 5. Extended the QFA mode to work with local files and/or
97 remote tapes and files. This way, restore can know in advance
98 the tape number and the offset for the inodes to extract and
99 can minimize the extraction time by seeking directly to the
102 6. Added the -A <archive> option to both dump and restore,
103 which makes dump to archive a dump table-of-contents in
104 the specified file to be used by restore to determine
105 whether a file is in the dump file that is being restored.
106 (the archive file syntax is also compatible with the
107 Solaris ufsdump generated one).
109 7. Small fix in restore making it able to read some (broken ?)
110 Solaris ufsdump tapes.
112 8. Fixed dump to correctly recognise the root filesystem when using
113 ext2 disk labels (LABEL=/). Thanks to John Yu <jky@it.bu.edu>
114 for reporting this bug.
116 9. Added the -P <file> option to restore to create a
117 Quick File Access file from an already made dump. Patch
118 contributed by Uwe Gohlke <uwe@ugsoft.de>.
120 10. Made restore compile and run on Solaris, making it a
121 possible replacement for the standard ufsrestore. Port was
122 contributed by Uwe Gohlke <uwe@ugsoft.de>.
124 Changes between versions 0.4b25 and 0.4b26 (released January 7, 2002)
125 =====================================================================
127 1. Added a set of backup scripts from Eugenio Diaz
128 <getnito@yahoo.com> in the examples section. It features
129 automatic (cron based) full and incremental dumping of
130 several filesystems on a separate filesystem.
132 2. Fixed a off-by-one miscalculation which disabled dumping
133 a one letter subdirectory. Thanks to Chris Siebenmann
134 <cks@utcc.utoronto.ca> for reporting the bug.
136 3. Fixed several restore bugs occuring when trying to
137 restore some missing files on the tape. Thanks to Chris
138 Siebenmann <cks@utcc.utoronto.ca> for reporting the bug.
140 4. Fixed --with-ldopts configure argument passing, installing from
141 a separate object directory, makefile cleanups contributed
142 by <splite@purdue.edu>.
144 5. Fix a bug which could caused, in some conditions, the highest
145 number inode of a filesystem, to not be dumped. Many thanks
146 to Chris Siebenmann <cks@utcc.utoronto.ca> for helping me
149 Changes between versions 0.4b24 and 0.4b25 (released November 17, 2001)
150 =======================================================================
152 1. Added a mini howto from Patrick Walsh in the examples
155 2. Minor man pages syntax corrections. Thanks to
156 Chris Danis <screechco@home.com> for reporting the bugs.
158 3. Added a script from David B. Peterson <dave@toppledwagon.com>
159 to the examples section. It features dumping several
160 filesystems to a remote tape drive upon ssh.
162 4. Added a patch provided by Richard Jones <rich@annexia.org>
163 which allows BRADEMAX (number of read errors tolerated by
164 dump) to be adjusted using the -I option.
166 5. Fixed a bug which disabled doing "restore -C -f -". Thanks
167 to Clemens Stiglechner <a9401816@unet.univie.ac.at> for the
170 6. Add the -l option to restore to specify if, when doing a
171 remote restore, the file used to access the backup is a
172 regular file (the defaults being a tape drive). Restore needs
173 to know this information when reading a remote compressed
174 dump. Previously, this information was autodetected, but
175 the autodetection code fails (with ioctl: Inappropriate ioctl
176 for device) when using a non Linux remote box. Thanks to
177 many users and especially to Eros Albertazzi
178 <eros@lamel.bo.cnr.it> for reporting this.
180 7. Found a workaroung for the dump deadlock problem (3 childs
181 stuck in pause(), father in read()). The workaround seems
182 to work for me and several beta-testers. If it doesn't work
183 for you, please report back.
185 8. Updated the RPM spec file (BuildPrereq, URL etc).
187 Changes between versions 0.4b23 and 0.4b24 (released September 12, 2001)
188 ========================================================================
190 1. Fixed the permissions of a newly created QFA file by dump.
192 2. Cleaned up the source of dump (the external variables
193 definition was a complete mess, making possible to have
196 3. Fixed restore to use the full tape volume path when doing
197 a compare (since it changes the working directory to the
198 filesystem being compared in the process).
200 4. Added the -q option to dump which makes dump abort
201 whenever it needs operator attention. It should be
202 easier to use dump in scripts with this option.
204 5. Detect the use of incompatible options to dump and
205 refuse them (like -a and -B options together).
207 6. Added bzip2 compression to dump/restore (use option -j level
208 to select it). Note that you will need a restore version
209 >= 0.4b24 in order to restore a bzip2 compressed dump.
210 The same warning as for the zlib compression applies:
211 the tape format of a bzip2 dump is not compatible with the
212 original BSD tape format.
214 7. Fixed a overflow problem in dump corrupting the dump when
215 very large files were encountered. Thanks to Vladimir Nadvornik
218 8. Added a ioctl(BLKFLSBUF) in dump which should flush the
219 kernel buffer/page cache before starting the dump, helping
220 a bit those who use dump on mounted filesystems. Thanks to
221 John Yu <jky@it.bu.edu> and to Theodore T'so <tytso@mit.edu>
224 9. Updated the RPM spec file following the RedHat changes
225 (dynamically linked binaries now in /usr/sbin etc).
227 10. Added a patch from Helmut Jarausch <jarausch@igpm.rwth-aachen.de>
228 which enables restore to recognise multi volume compressed dumps
229 done on CD. Included his scripts for dump (which pipe the dump
230 output directly into cdrecord) and restore. There is now
231 possible to to backups to CD on the fly!
233 Changes between versions 0.4b22 and 0.4b23 (released July 20, 2001)
234 ===================================================================
236 1. Fixed a buffer overflow in restore/tape.c. Patch provided by
237 Marcus Meissner (Caldera International Security Dept.).
239 2. Implement the Sun rmt extended protocol. Patch submitted
240 by Ian Gordon <iangordon@users.sourceforge.net>.
242 3. Check for the e2fsprogs header <ext2fs/ext2_fs.h> instead of
243 the linux kernel header. This ensures that dump always has the
244 latest version of this file. Patch submitted by
245 Andreas Dilger <adilger@turbolinux.com>.
247 4. Report any filesystem present in either /etc/fstab with a
248 positive passno or /etc/dumpdates in dump -w output.
249 Patch submitted by Andreas Dilger <adilger@turbolinux.com>.
251 5. Fixed the looping problem in dump introduced in the
254 6. Changed the -B option of dump to limit the size of
255 _compressed_ output per volume if compression is on.
256 Patch contributed by Helmut Jarausch
257 <jarausch@igpm.rwth-aachen.de>. Note however that, since
258 it is impossible to predict the size of the compressed
259 data before doing the compression, the -B limit is a bit
262 7. Fixed a bug in reading the operator typed file/tape path for
263 the next volume in restore.
265 8. Implemented a "-F script" option to restore which permits the
266 user to specify a script which will be launched at the
267 beginning of each tape, useful for automatic programming of
268 tape changers for example. See the restore man page for the
269 script parameters and return codes.
271 9. Small fix for the QFA routines provided by Uwe Gohlke
272 <uwe@ugsoft.de>, and some recommendations for QFA uses in
275 10. Fixed the multivolume restoring where making a mistake
276 to the 'Mount next tape' prompt caused several blocks to
279 11. Enhanced the -e option of dump to take as a parameter a
280 comma separated list of inode numbers.
282 12. Added the -E option to dump which specify a file containing
283 inode numbers to exclude from the dump.
285 13. Fixed the compressed multi-volume dump + restore.
287 Changes between versions 0.4b21 and 0.4b22 (released May 12, 2001)
288 ==================================================================
290 1. Made dump report the number of blocks dumped per volume.
291 Thanks to Kenneth Porter <shiva@well.com> for the suggestion.
293 2. Fix a bug in dump where waiting too much at the 'change volumes'
294 question caused the volume to be rewritten. Thanks to
295 Graham Reed <greed@users.sourceforge.net> for reporting the
296 bug and providing a patch.
298 3. Added a compression option to dump, contributed by Jerry
299 Peters <gapeters@worldnet.att.net>.
301 WARNING: the compression works only on files/pipes and
302 drives supporting variable block size.
304 WARNING2: tapes created using the compression feature are
305 incompatible with the standard dump tape format, and a
306 version of restore >= 0.4b22 will be needed for extraction.
308 4. Fixed some compilation problems for glibc 2.2.2 and 64 bit
309 architectures. Thanks to Paul Slootman <paul@debian.org> for
310 the patch and to Bdale Garbee <bdale@gag.com> for forwarding
313 5. Many cleanups (CPP defines, const char warnings, check of
314 ext2fs COMPAT flags, time_t cleanups, added libext2 version
315 in dump usage text) by Andreas Dilger <adilger@turbolinux.com>.
317 6. Made --prefix option work in configure. All the install path
318 are now based on the configure parameters.
320 7. Added the Quick File Access mode in dump/restore, contributed
321 by Uwe Gohlke <uwe@ugsoft.de>. In this mode, dump stores in
322 a file tape position for each inode, and this file is used by
323 restore (if called with parameter Q and the filename)
324 to directly position the tape at the file restore is currently
325 working on. This saves hours when restoring single files from
326 large backups, saves the tapes and the drive's head. Use
327 --enable-qfa option of configure to compile in the QFA support.
329 8. Added the possibility to dump several files and directories
330 in a single invocation of dump. Thanks to Uwe Gohlke
331 <uwe@ugsoft.de> for implementing this option.
333 9. Fixed the dumping and restoring of files with holes
334 (files ending with holes didn't get dumped or restored
337 10. Fixed a socket descriptor leak in dump, which leaved opened
338 3 file descriptors per dump process (and there is one dump
341 11. Fixed dump large file system support, by explicit use of
342 open64/lseek64/etc functions (explicit use needed because
343 e2fsprogs libraries don't behave well when compiled with
344 FILE_OFFSET_BITS=64).
346 Changes between versions 0.4b20 and 0.4b21 (released January 13, 2001)
347 ======================================================================
349 1. Fixed some bugs in the dump -w|-W logic introduced by
350 the previous version. Thanks to Andreas Dilger
351 <adilger@turbolinux.com> for his help on this one.
353 2. Fixed again a compilation problem when using old e2fs
354 headers (filesystem label related). Thanks to many users
355 who reported this stupid error.
357 3. Fixed a build problem on old lib5 based systems dealing with
358 _PATH_MNTTAB being undefined. Thanks to John Adams
359 <johna@onevista.com> for reporting the bug.
361 4. Improved the error detection in restore when creating
362 the temporary files in TMPDIR. Now restore will corectly
363 report a 'no space left on device' error instead of
364 strange errors which could imply an archive corruption.
365 Thanks to Gabriel Somlo <somlo@cs.colostate.edu> and
366 bgjenero <bgjenero@sympatico.ca> for reporting the bug.
368 5. Added the throughput information in dump's progression
369 messages. Thanks to Andreas Dilger <adilger@turbolinux.com>
372 6. Use libext2fs's inode scanning routines, which are particularly
373 robust when dealing with errors (especially when having some
374 bad blocks on the filesystem). Many thanks to Colin
375 <colin@horizon.com> for reporting the bug, investigating
376 the issues, finding a workaround, writing the patch and
377 fully testing it... (of course, if this patch will break
378 anything, he is to blame for :-)).
380 7. Made dump and restore LFS aware. Dump can dump a filesystem
381 containing large files, generate a large file on output and
382 restore can restore them. This feature is not enabled by
383 default, use --enable-largefile option of configure to enable
384 it (you need to have a LFS aware glibc though). Thanks to
385 Andrea Arcangeli <andrea@suse.de> for submitting the patch,
386 and to Theodore T'so <tytso@valinux.com> for his always
389 8. Made dump ask upon a tape write error if it should rewrite
390 the current volume (assume this is a bad tape) or if it should
391 assume an end-of-tape condition (useful for tape drives which
392 behaves badly at the end of the tape). Thanks to Andreas
393 Dilger <adilger@turbolinux.com> for the suggestion.
395 Changes between versions 0.4b19 and 0.4b20 (released November 10, 2000)
396 =======================================================================
398 1. Fixed a small compilation problem due to a change
399 in the definintion of the struct sigaction in
400 glibc 2.0 / libc5. Thanks to Gunther Schlegel
401 <schlegel@riege.de> for reporting the bug and to
402 Dave Platt <dplatt@snulbug.mtview.ca.us> for suggesting
405 2. Modified the label and uuid analysis in order to be
406 self-contained (independant of kernel/libc headers). This
407 should fix the compile with older kernel/libc headers and
408 will preserve the functionality. Thanks to Bernhard Erdmann
409 <bernhard.erdmann@gmx.de> for reporting the bug.
411 3. The 'exclude inode' option, if applied to a directory
412 inode, excludes now correctly all the leaves of this
413 directory. Thanks to John R. Dennison
414 <gerdesas@users.sourceforge.net> for reporting the bug.
416 4. Fixed the '-e' option to disable the possibility
417 to exclude the root inode (which causes the dumps to
418 be unreadable by restore). Prevented array overflow
419 when multiple -e options are used.
421 5. Fixed dump to correctly interpret a filesystem argument
422 which is a mountpoint and it is not an absolute path
423 (as specified in the fstab). Thanks to Bernhard R. Erdmann
424 <be@berdmann.de> for reporting the bug.
426 6. Made dump able to backup files larger than 2 GB. Note that
427 dump still doesn't cope with files larger than 4 GB.
429 7. Restore the real uid and gid before invoking an external
430 RSH program (big hole when dump or restore is suid root!).
432 8. Get the values for the default tape device and the location
433 of fstab file from the system headers. Thanks to
434 Andreas Dilger <adilger@turbolinux.com> for the patch.
436 9. Made dump -w|-W report all recognized filesystems
437 present in either /etc/fstab or /etc/dumpdates, and present
438 the list in the same order as in fstab file. Thanks
439 to Andreas Dilger <adilger@turbolinux.com> for the patch.
441 10. Made dump's -a (automatic end of tape detection) the
442 default. Specifying one of -B, -c, -d or -s options will
443 override the EOM detection. Thanks to Andreas Dilger
444 <adilger@turbolinux.com> for the patch.
446 11. Save the ext2 filesystem label into the dump volume label.
447 Specifying a volume label on the command line (-L option)
448 will override this feature. Thanks to Andreas Dilger
449 <adilger@turbolinux.com> for the patch.
451 Changes between versions 0.4b18 and 0.4b19 (released August 20, 2000)
452 =====================================================================
454 1. Fixed the signal handling in dump (which I broke in 0.4b17)
455 which was causing several strange dump failures (dump
456 hanged or segmentation faults etc).
458 2. Specified the default blocksize in the dump man page.
460 3. Changed two info messages of restore to be written on stdout
461 instead of stderr in order to leave stderr only for errors
462 or warnings. Thanks to Stephen Atwell
463 <satwell@urbana.css.mot.com> for the suggestion.
465 4. Corrected an off by one calculation which prevented
466 dumping a directory having a 1 character name length.
467 Thanks to Bernhard Erdmann <bernhard.erdmann@gmx.de>
468 for reporting the bug.
470 5. Reinforce dump in order to allow only level 0 and no
471 -u option when dumping a subdirectory, and document
472 this in the man page. Thanks to Bernhard Erdmann
473 <bernhard.erdmann@gmx.de> for reporting the bug.
475 6. Fixed a small harmless bug in restore which caused
476 in some conditions hard links to be restored several
477 times (therefore generation some warning messages).
478 Thanks to Kenneth Porter <shiva@well.com> for
481 7. Updated the RPM spec file to the latest RedHat version,
482 providing FHS packaging and other cosmetic changes.
483 You will need RPM version >= 3.0.5 in order to build the RPM.
485 8. Updated the configure script to check for libtermcap
486 before checking for libreadline, because we need this
487 library in order to compile the readline support.
489 9. Made dump understand the LABEL= and UUID= notation
490 both in /etc/fstab and on the command line. Note that
491 you will need the /proc filesystem in order to use
492 these notations. Thanks to Erik Troan <ewt@redhat.com>
493 for providing the patch.
495 Changes between versions 0.4b17 and 0.4b18 (released June 30, 2000)
496 ===================================================================
498 1. Fixed a potential buffer overflow in restore. Thanks
499 to Stan Bubrouski <satan@fastdial.net> for reporting
502 2. Fixed a readline-related bug which prevented
503 'cat DUMPFILE | restore -i -f -' from working. Thanks
504 to Charles Karney <karney@users.sourceforge.net>
507 3. Changed a few "panic" into "exit", causing restore to
508 be more stable against some attacks (like the last one
509 reported on Bugtraq, although the last version of restore
510 was not vulnerable - just dumped core). Thanks to
511 Andreas Hasenack <andreas@conectiva.com.br> for reporting
514 4. Removed the suid-root bit on dump and restore in the
515 default build (and generated RPMs). It should be safer
516 now. Users who need the suid-root capabilities in order
517 to make network backups should read first the man page
518 of dump and enable the suid bit manually.
520 5. Added -ltermcap to the compile parameters for restore
521 when using readline support, in order to make the compile
522 process work on some systems (SuSE ?). Thanks to
523 Patrik Schindler <poc@pocnet.net> for reporting the bug.
525 Changes between versions 0.4b16 and 0.4b17 (released June 1st, 2000)
526 ====================================================================
528 1. The -F script is called now *only* at the end of a tape,
529 not at the end of the dump. Thanks to Christian Haul
530 <haul@informatik.tu-darmstadt.de> for the bug report.
532 Normally, the device name and the current volume number
533 are passed on the command line of the script. However,
534 if you want the old style script (with no arguments
535 passed on the command line), you can enable it in
536 configure with the --enable-oldstylefscript.
538 2. Use posix signal handling to preserve dump functionality
539 with libc5. Thanks to Jeff Johnson <jbj@redhat.com> for
542 3. Made the exit code of restore in case of a 'C'ompare
543 command reflect the result of the compare. An exit status
544 of 0 means the dump archive is correct, 1 means tape errors,
545 2 means that some files were modified. Thanks to Kenneth Porter
546 <shiva@well.com> for the suggestion.
548 4. Made (finally) quotes work in interactive restore.
550 5. Small fixes in order to allow dump to compile with a
551 really old e2fsprogs version. Thanks to Ian Zimmerman
552 <itz@speakeasy.org> for the bug report.
554 6. Add GNU readline capabilities to interactive restore.
555 Use configure's --enable-readline flag to enable this feature.
556 Thanks to Patrik Schindler <poc@pocnet.net> for the
559 7. Do the compare on the fly in restore 'C' mode (this will
560 allow not to exhaust the available /tmp space when
561 ccmparing large files). Thanks to Kenneth Porter
562 <shiva@well.com> for the suggestion.
564 Changes between versions 0.4b15 and 0.4b16 (released March 11, 2000)
565 ====================================================================
567 1. Fixed some several duplicate 'DUMP: DUMP:' in the
570 2. Corrected the estimation of blocks to dump. Note that
571 this is still wrong for large directory inodes, due
572 to the size of a BSD directory entry on the tape
573 (estimation is based on the size of the ext2 directory,
574 which is smaller because it doesn't contain the null
575 character at the end).
577 3. Made dump report the total number of MB written to
578 tape at the end of run. Thanks to W. Reilly Cooley
579 <wcooley@nakedape.cc> for the patch.
581 4. Added the -X option to restore giving the possibility
582 to read the names of the files to be extracted/listed
583 from a text file (in addition of the command line).
584 Thanks to Dejan Muhamedagic <dejan@quant-x.com> for the
587 5. Added the device name and the current volume number
588 as arguments to the end of tape script (dump -F option).
590 6. Made the multi-volume dump work again (after having
591 broken it in 0.4b15).
593 Changes between versions 0.4b14 and 0.4b15 (released March 2, 2000)
594 ===================================================================
596 1. Added a prompt command in interactive restore mode. Thanks
597 to Andreas Dilger <adilger@home.com> for the patch.
599 2. Fixed a buffer overflow problem in dump (caused by
600 not checking the size of the filesystem parameter).
601 Thanks to Kim Yong-jun <loveyou@hackerslab.org> for
602 reporting this on Bugtraq (and to several dump users
603 who forwarded me his mail).
605 3. Added the '-F script' option to dump in order to
606 launch a script at the end of each tape (to be used
607 with a tape changer, or to notify the sysadmin by
610 4. Fixed a bug in restore compare code caused by the changes
613 5. Fixed the treatment of options using the old BSD syntax
614 in both dump and restore.
616 Changes between versions 0.4b13 and 0.4b14 (released February 10, 2000)
617 =======================================================================
619 1. Fixed a bug in dump which may have caused invalid deleted
620 directories to be dumped out if they were deleted after the
621 mapping pass. This could occure on active filesystem and lead
622 to heap corruption (causing dump malfunction in many possible ways).
623 Thanks to Jan Sanislo <oystr@cs.washington.edu> for finding this
624 bug and submitting the patch.
626 2. Fixed the handling of the filesystem parameter in dump. A
627 '/mnt/disk' parameter caused the disk contents to be dumped,
628 but a '/mnt/disk/' parameter caused the mountpoint directory
629 to be dumped (generally an empty directory).
631 3. Improved the output of dump in order to tell which directory
632 it is currently dumping (when dumping a subtree).
634 4. Added the '-e' exclude inode option to dump. Thanks to
635 Isaac Chuang <ike@isl.stanford.edu> for contributing with the patch.
637 5. Added a REPORTING-BUGS file in order to provide a guide
638 on how to correctly report a bug in dump/restore.
640 6. Fixed a restore problem when restoring a hard link to an inode
641 having the immutable or append-only attribute set. Thanks to
642 Ambrose Li <acli@mingpaoxpress.com> for submitting the patch.
644 7. Fixed a compatibility problem between dumps made on little
645 endian machines (the format was correct) and big endian
646 machines (the format was incorrect). This fix break the
647 compatibility with the older dumps made on big endian
648 machines (sparc, m86k, ppc etc). For the first time in
649 linux dump's history (I believe), the dumps made by *BSD,
650 Linux/alpha, Linux/sparc and Linux/x86 are compatible,
651 so interchangeable. Thanks to Rob Cermak
652 <cermak@ahab.rutgers.edu> for submitting the bug and
653 helping me test the fix.
655 8. Fixed the way dump reports the remaining percent/time, if
656 the number of blocks actually dumped exceeds the estimated
657 number of blocks. Thanks to Jean-Paul van der Jagt
658 <jeanpaul@dutepp0.et.tudelft.nl> for reporting the bug.
660 Changes between versions 0.4b12 and 0.4b13 (released January 21, 2000)
661 ======================================================================
663 1. Small Y2K fix in the man pages :). Thanks to Bernhard Sadlowski
664 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug.
666 2. Removed the requirement to build the RPM as root from the
667 spec file. Thanks to Christian Weisgerber
668 <naddy@mips.rhein-neckar.de> for submitting this.
670 3. Fixed a bug in dump related to the 'filetype' feature of ext2fs,
671 causing dump to block when dumping really huge filesystems.
672 Many thanks to Patrik Schindler <poc@pocnet.net> for
673 helping me find this bug.
675 4. Fixed the treatment for an interrupt signal when dump access
676 the remote tape through RSH. Thanks to Christian Weisgerber
677 <naddy@mips.rhein-neckar.de> for providing the patch.
679 5. Fixed a bug which was causing dump/restore to display
680 garbage characters instead of the remote host name.
682 Changes between versions 0.4b11 and 0.4b12 (released January 8, 2000)
683 =====================================================================
685 1. Small fix in the dump man page. Thanks to Thorsten Kukuk
686 <kukuk@suse.de> for submitting the patch.
688 2. Fix for the exit code when using the size estimate option of
689 dump. Thanks to Matti Taskinen <mkt@rni.helsinki.fi> for
690 submitting the patch.
692 3. Handle EINTR in atomical reads/writes in dump, which was causing
693 dump to fail on some systems. Thanks to Eric Jergensen
694 <eric@dvns.com> for reporting the bug and submitting the patch.
696 4. Handle more than 16 characters for the device names in dumpdates.
697 (up to 255 now). Thanks to Rainer Clasen <bj@ncc.cicely.de> for
698 tracking down the problem and proposing the solution.
700 5. Fixed a bug in dump which prevented the creation of the
701 dumpdates file when doing a 0-level dump without already
702 having a dumpdates file. Thanks to Patrik Schindler
703 <poc@pocnet.net> for reporting the bug.
705 6. Changed the way dump 'S' flag reports the size estimate
706 from number of blocks into bytes (making it compatible
707 with the Solaris version, and simplifying things for
708 amanda users). Thanks to Jason L Tibbitts III
709 <tibbs@math.uh.edu> for reporting the bug.
711 7. Fixed a compatibility problem in linux/alpha dump tape format.
712 Now the linux/alpha dump are (again) compatible with the
713 other dump formats. But this breaks compatibility with
714 older dumps made on alpha. Thanks to Mike Tibor
715 <tibor@lib.uaa.alaska.edu> for helping me in finding this bug.
717 Changes between versions 0.4b10 and 0.4b11 (released December 5, 1999)
718 ======================================================================
720 1. Added a '--enable-kerberos' to configure.
722 2. Added a 'S' option to dump which determines the amount of space
723 that is needed to perform the dump without actually doing it, similar
724 to the Sun's ufsdump 'S' option. Patch contributed by Rob Cermak
725 <cermak@ahab.rutgers.edu>.
727 3. Added a 'M' multi-volume option to dump and restore which enables
728 dumping to multiple files (useful when dumping to an ext2
729 partition to make several dump files in order to bypass the 2GB
730 file size limitation). The 'f' argument is treated as a prefix and
731 the output files will be named <prefix>001, <prefix>002 etc. With
732 the 'M' flag, restore automatically selects the right file without
733 asking to enter a new tape each time.
735 4. Fixed a memory leak which was causing dump to grow very big
736 (270MB when dumping a 10GB filesystem...). Thanks to Jason
737 Fearon <jasonf@netrider.org.au> for reporting the bug.
739 Changes between versions 0.4b9 and 0.4b10 (released November 21, 1999)
740 ======================================================================
742 1. Make configure test if the system glob routines support
743 extended syntax (ALTDIRFUNC). If not, use the internal glob
744 routines instead of system ones. Thanks to Bernhard Sadlowski
745 <sadlowsk@Mathematik.Uni-Bielefeld.DE> for reporting the bug
746 and helping me resolve this and other minor libc5 compiling
749 2. Fix a problem when dumping a ext2fs with the 'filetype'
750 feature enabled. Thanks to Patrick J. LoPresti
751 <patl@cag.lcs.mit.edu> for reporting the bug and to
752 Theodore Y. Ts'o <tytso@mit.edu> for providing the patch.
754 3. Made the nodump flag work on directories. A directory which
755 has the nodump flag gets never dumped, regardless of its
758 4. Integrate a patch from Jeremy Fitzhardinge <jeremy@goop.org>
759 which allows dump on an active ext3 filesystem. However, this
760 is a "quick and dirty" patch which enables backup of an ext3
761 filesystem through the ext2 compatibility (by ignoring the
762 NEEDS_RECOVERY bit). The journal file is not recognized and
763 it is dumped (it should not).
765 5. Test the superblock compatibility flags when dumping, in order
766 to be sure that we know how to deal with specific features.
768 Changes between versions 0.4b8 and 0.4b9 (released November 5, 1999)
769 ====================================================================
771 1. Use lchown instead of chown, fixing a possible security problem
772 when restoring symlinks (a malicious user could use this
773 to deliberately corrupt the ownership of important system files).
774 Thanks to Chris Siebenmann <cks@utcc.utoronto.ca> for detecting
775 this and providing the patch.
777 Changes between versions 0.4b7 and 0.4b8 (released November 3, 1999)
778 ====================================================================
780 1. Put dump sources under CVS, added Id tags in all files so
781 one can use 'ident' on binary files.
783 2. Added the dump/restore version in the usage text so one can
784 easily verify the version he is using.
786 3. Small patch from Nuno Oliveira <nuno@eq.uc.pt> which fixes
787 a va_start/va_end problem on linux-ppc (always call va_start
788 va_end in pairs each time we use a vararg function).
790 4. Added again the DT_* constants because old libc does not
791 contain them :(. Thanks to Eric Maisonobe <virnet@nat.fr>
792 for submitting the bug report.
794 5. Use ext2fs_llseek instead of llseek. With recent e2fsprogs
795 this should enable dumping big (huge) filesystems.
797 6. Added the RSH environment variable in order to be able to
798 use a rsh replacement like ssh when doing remote backups (and
799 bypass the security limitations of rcmd). Now you can do remote
800 backups without being root (or making dump setuid root).
802 7. Modified again the way dumpdates works. For incremental dumps,
803 we need to read dumpdates even if we are not using 'u' option.
804 Thanks to Bdale Garbee <bdale@gag.com> for his ideas on how
807 Changes between versions 0.4b6 and 0.4b7 (released October 8, 1999)
808 ===================================================================
810 1. Removed the 'k' flag from the restore 'about' text if kerberos
813 2. Prototyped (f)setflags from e2fsprogs and corrected the calls
814 to them (fsetflags takes a char*, setflags an open fd!).
816 3. (f)setflags is called only if the flags aren't empty. If the
817 file is a special file, a warning is printed, because changing
818 flags implies opening the device. Normally, a special file
819 should not have any flag... (Debian bug #29775, patch provided
820 by Abhijit Dasgupta <abhijit@ans.net>).
822 4. Made possible to dump a file system not mentioned in /etc/fstab.
823 (Debian bug #11904, patch provided by Eirik Fuller <eirik@netcom.com>).
825 5. Changed the default behaviour to not create dumpdates
826 unless 'u' option is specified. Removed the old "debian-patch"
827 which provided the same thing. (Debian bug #38136, #33818).
829 6. Removed all those dump*announce, since they were getting old...
831 7. Added warning messages if dumpdates does not exist and
832 when an inferior level dump does not exist (except for a level 0
835 8. Debugged the glob calls in interactive mode: restore used a
836 dirent struct which was different from the /usr/include/dirent.h
837 one (this used to work, is it a glibc2 change?), so none of the
838 compat glob (which used /usr/include/dirent.h) or the system glob
839 worked. Restore use now the system dirent (and the system
840 DT_* constants), which are compatible with BSD ones.
842 9. Added a configure flag (--with-dumpdatespath) to specify
843 the location of dumpdates file. By default, it is
846 10. Added the "AUTHOR" and "AVAILABILITY" sections and
847 included the current date/version in man pages.
849 11. Corrected the estimation of remaining time when
850 the operator doesn't change the tapes quickly enough. This
851 was an old bug, I thought I corrected it, and discovered
852 that in fact it was corrected in two different places, so
853 the results canceled each other...
855 Changes between versions 0.4b5 and 0.4b6 (released October 1, 1999)
856 ===================================================================
858 1. Integrated multiple patches from RedHat, Debian and SuSE:
860 - tweak dump/itime.c to not try to read dumpdates if the 'u' option
862 - several fixes in the man pages.
863 - update the default tape device to /dev/st0.
864 - many updates for Linux Alpha (byte ordering, size_t etc).
866 - use environment variable for TMPDIR (instead of /tmp).
867 - use sigjmp_buf instead of jmp_buf (RedHat bug #3260).
868 - workaround egcs bug (RedHat bugs #4281 and #2989).
869 - wire $(OPT) throughout Makefile's.
871 2. Upgrade the dump revision to 1, making possible to dump filesystems
872 made with e2fsprogs-1.15 or newer. Nothing seems to break...
874 3. Fix some compile warnings, prototype all functions.
876 4. Use glibc err/glob instead of internal compatibility
877 routines (only if available).
879 5. Fix a compile error on Linux 2.2.7 / libc5 (5.4.44) (patch provided
880 by Bernhard Sadlowski <sadlowsk@mathematik.uni-bielefeld.de>).
882 Changes between versions 0.4b4 and 0.4b5 (released September 22, 1999)
883 ======================================================================
885 1. Integrated the changes from FreeBSD-3.1-RELEASE
886 (mostly bug fixes, buffer overruns, dump has now an "automatic
887 tape length calculation" flag, dump/restore can use kerberos now
888 (this is NOT tested), use environment variables for TAPE and
891 2. Integrated three RedHat patches ("glibc", "kernel" and "bread" patches)
893 3. Corrected a bug in restore when using 'C' option with multi-volumes
894 tapes (files splited accros two tapes give "size changed" errors
897 4. Corrected the long standing bug when dumping multiple tapes.
898 This works for me, needs further testing.
900 Changes between versions 0.4b3 and 0.4b4 (released January 17, 1997)
901 ====================================================================
903 1. Dump now runs correctly on kernels 2.1.x
904 Fix made by Gerald Peters <gapeters@worldnet.att.net>
906 Changes between versions 0.4b2 and 0.4b3
907 ========================================
909 1. Use realpath() if available
913 Changes between versions 0.4b1 and 0.4b2
914 ========================================
916 1. Fixed the bug fix from Greg Lutz (I had made a mistake when integrating
919 2. Fixed restore to make it able to read FreeBSD 2.x dumps again
921 3. Fixed configure.in to correctly handle --enable-rmt
923 Changes between versions 0.3 and 0.4b1
924 ======================================
926 1. Integrated the changes from 4.4BSD-Lite2
928 2. Integrated the patches from Debian and Red Hat
930 3. Portability changes: use the __u32, __u16, __s32, and __s16 types
932 4. Changed dump to use the Ext2fs library to get block addresses. This
933 should solve the endianness problem on SparcLinux.
935 5. Created a configure.in file (shamelessly stolen from the e2fsprogs
936 distribution's one) to use autoconf
938 6. Fixed a few minor bugs
940 Changes between versions 0.2e and 0.2f
941 ======================================
943 1. Added the creation of named pipes (fifos) in restore.
945 2. Added the -N flag in the restore manual page.
947 3. Added the file kernel.patch which contains the llseek() optimization
948 patch for 1.2.x kernels.
950 4. Fixed a bug in the restoration of symbolic links: owner and group were
953 5. Integrated some changes from FreeBSD 2.2.
955 6. Added a call to ftruncate() after restoring each file to restore
956 correctly files ending by a hole.
958 Changes between versions 0.2d and 0.2e
959 ======================================
961 1. Fixed a bug in the "set owner/mode" process. Every file was restored
962 with owner = root (0) and group = root/wheel/whatever (0).
964 Changes between versions 0.2c and 0.2d
965 ======================================
967 1. Dump is now able to backup 2GB+ filesystems.
969 2. Dump and restore can now be linked as static binaries.
971 Changes between versions 0.2b and 0.2c
972 ======================================
974 1. Fixed a bug when dumping ``slow'' (i.e. normal) symbolic links.
976 Changes between versions 0.2a and 0.2b
977 ======================================
979 1. Really fixed the bug that I should have corrected in 0.2a.
981 2. Enabled optimization again.
983 Changes between versions 0.2 and 0.2a
984 =====================================
986 1. Disabled the optimization during compilation.
988 Changes between versions 0.1 and 0.2
989 ====================================
991 1. Fixed a bug in fstab.c which caused a null pointer to be stored in
992 the fs_type field (actually, I modified the file fstab.c to make it
993 use the mntent functions).
995 2. Dump and restore now use a 4.3 BSD compatible dump format. Backups
996 made by dump should be readable by the BSD restore and backups made
997 by the BSD dump should be readable by restore. Unfortunately, this
998 means that the dump format has changed between version 0.1 and version
1001 3. Dump is now able to backup a subtree, it is no longer limited to whole
1002 filesystems like the BSD version.
1004 4. Dump now uses ext2_llseek() so it is able to backup filesystems bigger
1007 Changes between versions 0.0 and 0.1
1008 ====================================
1010 1. Now create links rdump and rrestore during the `make install' step.
1012 2. Linux port specific bugs added to the manual pages
1014 3. Incorrect estimation of the number of tapes blocks fixed when doing
1015 incremental backups.
1017 4. Better ls-like format in restore in interactive mode.